Building The Somali National Army Anatomy Of A Failure 2008 2018

Journal of Strategic Studies

ISSN: 0140-2390 (Print) 1743-937X (Online) Journal homepage: https://www.tandfonline.com/loi/fiss20

R

Routledge

Taylor & Francis Croup

Building the Somali National Army: Anatomy of a
failure, 2008-2018

Paul D. Williams

To cite this article: Paul D. Williams (2019): Building the Somali National Army: Anatomy of a
failure, 2008-2018, Journal of Strategic Studies, DOI: 10.1080/01402390.2019.1575210

To link to this article: https://doi.Org/10.1080/01402390.2019.1575210

Published online: 27 Feb 2019.

B*

Submit your article to this journal C?

Article views: 168

CrossMark

View Crossmark data G?

Full Terms & Conditions of access and use can be found at
https://www.tandfonline.com/action/journal I nformation?journalCode=fjss20

THE JOURNAL OF STRATEGIC STUDIES
https://d 0 i. 0 rg/l 0.1080/01402390.2019.1575210

O Routledge

Taylor & Francis Group

ARTICLE

Check for updates

Building the Somali National Army: Anatomy of
a failure, 2008-2018

Paul D. Williams 0

Elliott School of International Affairs, TheGeorge Washington University, Washington, DC,
USA

ABSTRACT

Over a decade of security force assistance (SFA) initiatives to build an effective
Somali National Army (SNA) failed because of the interrelated effects of political,
contextual and operational challenges. The key political challenges were interest
asymmetry between international actors and Somali elites, insufficient focus on
institution-building and a lack of donor coordination. The principal contextual
challenges in Somalia were the legacies of two decades of state collapse and
the negative effects of clan dynamics. The main operational challenges were
building an army while simultaneously fighting a war, the complexities of
military integration, and the severe capability gaps afflicting the SNA.

KEYWORDS Security force assistance (SFA); security sector reform (SSR); Somalia; African Union; al-
Shabaab

Introduction

In December 2017, Somali government officials delivered the results of an
Operational Readiness Assessment of the Somali National Army (SNA) to
a conference of international partners. 1 The assessment was needed
because neither the government nor its partners had reliable basic informa¬
tion about the army, including the identities of SNA personnel, their loca¬
tions and unit affiliations, or their weapons and equipment. The assessment
confirmed the SNA was in a dire state. 2 There were fewer frontline personnel
than previously estimated (on average battalions had only 63% of their
authorised strength), there were inconsistent recruiting standards (for both
officers and rank-and-file troops), and most battalions lacked basic

CONTACT Paul D. Williams Q pauldw@gwu.edu Q George Washington University
'in May 2017, the Federal Government of Somalia (FGS) recognised the value of conducting 'a Joint
Verification exercise' of its armed forces. SNA presentation to AMISOM Joint Review team,
Mogadishu, 23 May 2017. Following discussions with its international partners, the FGS agreed to
conduct a comprehensive verification of the SNA and Somali Police Force across all sectors of
operations as part of AMISOM's exit strategy. AU document PSC/PR/COM.(DCC), 12 July 2017, §15.
This exercise was subsequently renamed the Operational Readiness Assessment, which began in
September.

2 Operational Readiness Assessment - Final Report (Somali Ministry of Defense 1 December 2017).

© 2019 Informa UK Limited, trading as Taylor & Francis Group

2 @ P. D. WILLIAMS

equipment, including weapons, ammunition, communications kit and vehi¬
cles. The result was an army in name only, largely confined to defensive and
localised operations, unable to undertake a coherent national campaign,
and often reliant on the African Union Mission in Somalia (AMISOM) and
United Nations (UN) for protection, securing its main supply routes, logistics
support and casualty evacuation.

The embarrassing reality was that after a decade of receiving security
force assistance (SFA) 3 from over a dozen 4 international actors, only one
SNA unit was able to conduct sustained offensive operations; an advanced
infantry battalion known as Danab ('lightning') which was trained, equipped
and mentored by the United States and kept largely separate from the rest
of the army. 5 This wasn't for a lack of trying. For example, part of AMISOM's
original mandate in 2007 was to assist with 'the effective re-establishment
and training of all-inclusive Somali security forces,' some of which occurred
in Uganda. 6 To give some examples of the types and scale of SFA to the
SNA, outside of AMISOM, by late 2008 Ethiopia had reportedly trained
17,000 Somali security forces 7 ; since 2010, the European Union Training
Mission (EUTM) trained about 5700 SNA, at a cost of approximately
€90 million 8 ; from November 2013 the UN Support Office for AMISOM
(UNSOA) provided logistical support for up to 10,900 SNA troops engaged
in joint operations with AMISOM 9 ; the UAE deployed about 200 advisers in
Somalia and trained a brigade of soldiers in Mogadishu as well as providing
additional stipends to others; since 2009, the US State Department and
Pentagon both obligated SFA monies to the SNA, including for tactical
engagement for military operations, logistics, medical support and commu¬
nications capabilities, and institutional reform including civilian control of
the military 10 ; between 2010 and 2013, Italy paid stipends for approximately
3300 SNA troops 11 ; since 2012, the UK deployed a mission support team of

3 The United States military defines SFA as 'unified action to generate, employ, and sustain local, host
nation or regional security forces in support of a legitimate authority'. US, Commander's Handbook for
Security Force Assistance (Kansas: Joint Center for International Security Force Assistance
14 July 2008), 1.

4 The providers of SFA to the SNA since 2008 include Burundi, Djibouti, Ethiopia, Italy, Kenya, Sudan,
Turkey, the UAE, Uganda, the United Kingdom, the United States, as well as the African Union (AU),
European Union (EU) and United Nations (UN).

5 The name came from a commando battalion used by the SNA before the civil war of the late 1980s.

6 S/RES/1744, 20 February 2007, §4.

7 Report of the Monitoring Group on Somalia pursuant to Security Council resolution 1811 (2008) (S/2008/
769, 10 December 2008), §19.

8 EUTM website, https://www.eutm-somalia.eu.

9 S/RES/2124, 12 November 2013.

10 Between 2010 and May 2017, the US obligated approximately $66 million for SNA stipend-related
activities. Inspection of the Bureau of African Affairs' Foreign Assistance Program Management (US
Office of Inspections, ISP-l-18—02, October 2017), 11.

11 Report of the Monitoring Group on Somalia pursuant to Security Council resolution 1853 (2008) (S/2010/
91, 10 March 2010), §191 and Colin Robinson, 'Revisiting the rise and fall of the Somali Armed Forces,
1960-2012', Defense & Security Analysis 32/3 (2016), 249.

THE JOURNAL OF STRATEGIC STUDIES @ 3

about 70 troops to strengthen the SNA and Somali Ministry of Defence 12 ;
and from 2015 to 2017, Turkey spent some $50 million building new training
facilities for the SNA in Mogadishu, provided direct budget support to the
Federal Government, some of which was reportedly used to pay salaries,
and deployed approximately 200 of its own troops to train an SNA battalion
each year. 13

This article argues that the collective failure of these initiatives stemmed
from the interrelated negative effects of political, contextual and operational
challenges in Somalia. The key political challenges were interest asymmetry
between international actors and local Somali elites, insufficient focus on
institution-building, and a lack of donor coordination. The principal contextual
challenges in Somalia were the legacies of two decades of state collapse and
the negative effects of clan dynamics. Finally, the main operational challenges
involved building an army while simultaneously fighting a war, the complex¬
ities of military integration, and the huge capability gaps afflicting the SNA.

To substantiate this argument, the article begins by summarising the evolu¬
tion of Somalia's armed forces since the country's independence, focusing on
the period since 2008 when the basis of today's SNA was established. Since
then, the Somali armed forces have remained fragmented and their institutions
and structures largely dysfunctional. The three subsequent sections then ana¬
lyse the main political, contextual and operational challenges that prevented
international partners from building an effective SNA. The conclusion reflects
on the relationship between these different problems and briefly proposes
some alternative options. Evidence is drawn from interviews with relevant
experts as well as previously unpublished material about the state of the SNA
gathered from the Somali government and its international partners.

The evolution of the Somali armed forces

By the late 2000s, there were two principal reasons to build an effective Somali
army. First, it was considered a vital part of resurrecting a functioning Somali
state that could stabilise the country and protect its citizens, first and foremost
from one of the world's deadliest militant groups, Harakat al-Shabaab. 14
Second, it became a crucial part of a viable exit strategy for AMISOM. 15 This
international project generated a long list of aspirational adjectives to make the

12 For details of the UK's support see Jon Lunn, Somalia: February 2017 Update (House of Commons
Briefing Paper No.7298, 16 February 2017).

13 See Robinson, 'Revisiting', 247 and Menekse Tokyay, 'Turkish military base in Somalia', Arab News,
17 August 2017, http://www.arabnews.com/node/1145846/middle-east.

14 Although debate continues over whether Somaliland will reunify with the rest of Somalia, this article
focuses on developments in south-central Somalia and to a lesser extent the autonomous region of
Puntland.

15 Paul D. Williams with Abdirashid Hashi, Exit Strategy Challenges for the AU Mission in Somalia
(Mogadishu: HIPS, February 2016).

4 @ P. D. WILLIAMS

new SNA legitimate, inclusive, cohesive, professional, sustainable, and effective.
Following the London Security Pact agreed in May 2017, official communica¬
tions from the Federal Government of Somalia (FGS) and its international
partners settled on the alliterative language of building security forces that
were affordable, acceptable, accountable, and able to protect the Somali
population. 16 Part of the reason why building this new SNA proved so difficult
was the complicated evolution of its previous iterations.

While Somalia established a police force in 1943, its Armed Forces were not
formed until April 1960 by the merger of the Italian and British mandates'
mobile security forces. Initially, these numbered between 2000 and 5000 -
personnel. 17 For most of the next two decades, the Armed Forces focused on
pursuing an irredentist agenda designed to unite the Somali populations
spread across the wider Horn of Africa. Ultimately, this led to the decision to
invade Ethiopia in 1977, resulting in Somalia's defeat within less than a year.
During the 1980s, the Armed Forces became increasingly politicised and its
officer corps distorted by President Siad Barre's promotion of officers from his
preferred clans, mainly Marehan and related Darod. By the time Somalia's civil
war got underway in 1987, the Armed Forces had effectively started disinte¬
grating as Barre lost control of large parts of the country. 18

The Somali army subsequently ceased to exist until the new transitional
Somali authorities tried to resurrect it between 2000 and 2005. The first
effort was made in Djibouti with the establishment of the Transitional
National Government (TNG) in 2000, which recruited freelance gunmen
and members of Siad Barre's military. 19 Some of these carried over to 2004
when the Transitional Federal Government (TFG) was formed in Kenya and
then moved its military to the Somali town of Jowhar. They comprised of
some troops from Barre's military but also militias from Puntland and other
clans from the Jowhar area. When the TFG moved to Baidoa in 2005 it also
recruited some clan militias from the Bay region. 20

On paper at least, the SNA was part of the Somali National Security and
Stabilization Plan (NSSP), which the TFG adopted in June 2006. 21 The NSSP was
intended 'to serve as the main conduit for supporting collaboration between
the TFG and its international partners for sustained, coherent, harmonized and
complimentary security sector stabilization in Somalia.' 22 It was subsequently

16 Communique of the London Conference on Somalia, 11 May 2017, https://www.gov.uk/government/
publications/london-somalia-conference-2017-communique .

17 Robinson, 'Revisiting', 239.

18 lbid„ 241.

19 Mohamed Mubarak, 'Somali Military has more Problems than Lack of Guns', African Arguments,
28 February 2014, http://africanarguments.org/2014/02/26/somali-military-has-more-problems-than-
lack-of-guns-by-mohamed-mubarak/ .

20 lbid.

21 Report of the Chairperson of the Commission on the Situation in Somalia (AU document P SC/P R/2 (LX IX)
19 January 2007), §33.

22 Cited in the President's Foreword to the 2011 Somali NSSP.

THE JOURNAL OF STRATEGIC STUDIES @ 5

revised several times because it was originally intended to serve as guidance
during the transitional governance period. Revisions were developed in light of
recommendations made by two security sector assessments conducted during
2009 and 2011. 23 Rather ominously, the 2009 assessment concluded that:
'Security Sector Reform (SSR) efforts will probably falter - and quite possibly
even fail - without a broader political foundation that promotes good govern¬
ance, transparency and accountability within the Transitional timeframe.' 24

The basis of today's SNA, however, lay in the formation of a new Joint Security
Force in 2008. This comprised of security forces from both the TFG and the
Alliance for the Re-Liberation of Somalia (ARS)-Djibouti faction as stipulated in
the Djibouti Agreement of August 2008. This joint force probably numbered just
over 5,000 troops; about 3,000 from the TFG and about 2,000 from the ARS-
Djibouti, who were mostly militia from the Islamic Courts that seized Mogadishu
from the ruling warlords in mid-2006. It was to be given UN support to build their
capacity to restore the security sector and the rule of law. 25

Since then, it proved very difficult to ascertain the precise number of
personnel in the SNA. In January 2009, for instance, AMISOM's Force
Commander noted that both the TFG and ARS had submitted lists of names
of soldiers to AMISOM to be placed into the Joint Security Force but his troops
had to scrutinise the lists of names 'one by one' to verify each individual's
information and then issue them an ID card. Seven months into the process,
AMISOM had issued only approximately 800 ID cards. 26 It was also difficult to
distinguish genuine 'fighters' from other 'freelancers' who were, as the UN
Monitoring Group put it, "'on call" to varying degrees if required though always
ready to collect salary and stipend payments when available.' 27 Table 1 pro¬
vides various estimates of the SNA's strength from 1963 to 2017.

By early 2009, the Joint Security Force held about half of Mogadishu's 16
districts. Arguably its most functional unit was the TFG's 867-strong presiden¬
tial guard, based at the presidential compounds in Mogadishu and Baidoa 28
The President also controlled the Darwish; traditionally a composite force
carrying out military and police functions that generally operated indepen¬
dent from the SNA and answered directly to the President. 29

23 The 2009 assessment was compiled outside of Somalia, largely donor-led and based on interviews
with key actors. It was limited to south-central Somalia for political reasons and was stimulated by
donor concerns about the TFG forces. The 2011 assessment was carried out in Somalia and was more
Somali-driven and linked to the idea of developing a revised NSSP. Author's communication,
contractor engaged in the process, 31 July 2012.

‘‘''Somalia Security Sector Assessment (AU, US, EU, TFG, World Bank and UN: 1 January 2010), §1 i.

25 'Letter from the Secretary-General to the President of the Security Council' (S/2008/804,
19 December 2008), Annex.

26 Wikileak Cable, 09ADDISABABA717, 25 March 2009, §8.

27 Report of the Monitoring Group on Somalia and Eritrea pursuant to Security Council Resolution 2244
(2015) (S/2016/919, 31 October 2016), Annex 2.1 §3.

28 Robinson, 'Revisiting', 243.

29 Somalia Security Sector Assessment (January 2010), §16a9b12b.

6 @ P. D. WILLIAMS

Table 1 Estimated size of the Somali armed forces, 1963-2017.

Date

Estimated Size

Source

1963

4000

1965

4600

1970

12,000

1977

35,000-37,000

May 2009

5000-6700

TFG and ARS-Djibouti

November 2009

6400

(+ 6270 aligned forces)

UN Monitoring Group

November 2010

8000

AMISOM (CONOPS)

May 2011

10,106

TFG Security Officials

November 2011

10,299

AMISOM

February 2012

11,200

AMISOM (CONOPS)

September 2013

20,000

(inc. aligned militia)

UN Secretary-General

October 2013

12,000

Roger Carstens (contractor)

January 2014

22,714

FGSand UN

January 2017

19,440

UN and World Bank

March 2017

24,684

FGS and UK

March 2017

Around 20,000

(inc. aligned militia)

AU-UN Information Support Team

May 2017

24,820

(+4,569 benefit recipients)

FGS

Source: Paul D. Williams, Fighting for Peace in Somalia: A history and analysis of the African Union
Mission, 2007-2017 (Oxford: Oxford University Press, 2018), 242.

In June 2009, the TFG signed a cooperation agreement with the
Ethiopian-backed Ahlu Sunna Wal Jama'a (ASWJ) whereby the TFG would
provide them with military and logistical support and consult it on admin¬
istrative issues in areas under its control. The ASWJ reportedly had about
10,000-12,000 fighters, including 3,000 in Mogadishu. 30

In November 2009, the UN Monitoring Group estimated there were 2,900
effective TFG armed forces, with roughly 3,500 additional troops unvetted,
untrained and unregistered. 31 An additional 6,270 government-aligned troops
were estimated to exist, mostly clan militia and old, so-called 'grey' soldiers—
would-be pensioners who retained their uniforms from the Siad Barre era. 32

By May 2011, a second security sector assessment estimated there were
now 10,106 personnel in the SNA, excluding friendly militias such as ASWJ. 33
It also noted high levels of nepotism, corruption and poor record-keeping
and emphasised the urgent need for a biometric identification system across
the entire Somali security sector. Rather optimistically, in August 2011, the
UN Secretary-General concluded 'the biometric identification system
initiated in January should cover the entire National Security Force in
Mogadishu by October.' 34 It was still incomplete in late 2018.

30 lbid, §16a15.

31 S£MG Report, 10 March 2010, §17.

32 Somalia Security Sector Assessment (January 2010), §16a9a.

33 Security Sector Assessment (June 2011), 5.

34 Report of the Secretary-General on Somalia (UN document S/2011/549, 30 August 2011), §69.

THE JOURNAL OF STRATEGIC STUDIES @ 7

By the end of 2011, AMISOM estimated that there were about 10,300 SNA
troops organised into five brigades, with an additional 10,000 fighters in
various aligned militia. The AU concluded they were 'a fragmented disparate
group' whose 'strong prevalence of clan identity ... is the stumbling block
to all efforts in organizing a national military force.' 35 Moreover, the AU
observed that some of 'those who defect from Al Shabab are absorbed into
the TFG without undergoing serious scrutiny and rehabilitation.' 36 A few
months later, AMISOM estimated there were 11,200 SNA operating in
Mogadishu and another 11,000 (including ASWJ) engaged in operations
against al-Shabaab elsewhere. 'None of these forces,' AMISOM concluded,
'have the ability to support and sustain themselves logistically.' 37

In November 2013, the UN Support Office for AMISOM (UNSOA) was
tasked with providing non-lethal support to 10,900 SNA personnel engaged
in joint operations with AMISOM. 38 Consisting of food, water, fuel, transport,
tents, and in-theatre medical evacuation, this support was funded by the UN
Trust Fund for AMISOM and the Somali National Army, not the UN's assessed
peacekeeping contributions (like most of UNSOA's activities). UNSOA's sup¬
port to the SNA was supposed to conform to the UN's Fluman Rights Due
Diligence Policy for supporting non-UN security forces but there were huge
obstacles to verifying compliance.

By mid-2014, a sixth SNA brigade had been established. 39 As more attention
focused on supporting the SNA to coordinate operations with AMISOM, interna¬
tional partners and the Federal Government developed the Guulwade ('victory'
plan). 40 Launched in March 2015, this was intended to revamp the SNA with
international support but also provide some structure for the roughly 10,000 SNA
members that were not part of the UNSOA support package. The Guulwade plan
recognised that the SNA was little more than a collection of clan militias without
a functioning, centralised command and control structure. It argued that in order
to fight effectively the SNA needed better equipment, infrastructure, organisa¬
tion, and morale. In the short-term, it also badly needed increased mobility and
firepower to conduct effective operations. In essence therefore, the new 'plan'
was actually a list of capability gaps that the SNA wanted to fill, with rough
calculations of how much it would take to do it. The estimated costs were just
under $85 million for equipment, vehicles, small arms and light weapons, training
centres, medical facilities and barracks but excluding operations and mainte¬
nance costs. 41 This figure is notable for being well under half the amount the

35 Report on TFG Military Forces (November 2011) (AU internal document, November 2011), 1.

36 lbid, 7.

37 Military Strategic CONOPS (February 2012), §15.

38 See Paul D. Williams, Fighting for Peace in Somalia: A Flistory and Analysis of the African Union Mission
(AMISOM), 2007-2017 (Oxford UP, 2018) Ch.8.

39 See UNSOM, The Somali National Army (UNSOM briefing Version 1.06, 31 January 2015).

40 ln June, the Federal Government launched its Heegan ('readiness') plan for the police force.

41 Guulwade (Victory) Plan (SNA internal document, 9 April 2015), 9.

8 @ P. D. WILLIAMS

United States alone gave to Somalia in SFA during 2015/16. 42 Some critics of the
plan pointed to the clan composition of the 10,900 SNA, which was dominated
by one of the Hawiye sub-clans, in large part because it drew most of its recruits
from Hawiye-dominated areas around the Banadir region. Critics suspected
Hawiye leaders pushed the plan for precisely this reason. 43 There were also
complaints from within the Somali government that some international partners
pushed for the plan to focus too narrowly on the SNA troops that would be
supported by UNSOA rather than the entire force. 44

Table 2 shows the distribution of the designated 10,900 SNA that would
receive UNSOA support. It was particularly notable that any SNA in Sector 2
would be subject to consultation with the Interim Jubba Administration
largely because its leader, Ahmed Madobe, had come to power through
the Ras Kamboni militia and used them as his principal security force.

By the time of the 2017 London conference on Somalia—the principal
meeting of more than 40 of Somalia's key international partners—the SNA
was estimated at approximately 2100 officers, 18,700 soldiers and NCOs, as
well as nearly 3900 orphans, disabled and retirees. 45 It was organised into 12
brigades, over 30 battalions, and had various special units including the
presidential guard, military police, the Danab advanced infantry, and the
health unit. However, as noted at the start of this article, the Operational
Readiness Assessment in late 2017 revealed that most SNA battalions were
operating at only about two-thirds of their official strength. In December 2017,
the World Bank estimated that recurrent annual costs to implement Somalia's
new national security architecture would cost approximately $140 million. 46

As the only fully functioning unit of the SNA, it is worth briefly summar¬
ising Danab's origins in order to highlight why it remained the exception.

Table 2 Distribution of 10,900 SNA conducting joint operations with AMISOM (2015).

Region

AMISOM

Sector

Human Rights
Trained
(as at 9 April)

Guulwade

Plan

Human Rights Training
Required

Banadir, Lower/Middle

1 & 5

3139

3000

0

Shabelle

Bay, Bakool, Gedo

3

3495

2570

0

Hiraan, Galgaduud

4

2137

2000

0

Middle/Lower Jubba,

2/Kismayo

0

2880

2880

Gedo

Danab Battalion

Baledogle

300

450

150

Totals

8771

10,900

3030

Source: Guulwade (Victory) Plan (SNA internal document, 9 April 2015), 1.

^Communication, US official, 10 January 2018.

^Communication, UN official, 1 September 2017.

'■’interview, Somali government official, Djibouti, 27 February 2016.

45 Securing Payroli: Somali National Army (FGS and DFID, unpublished document, March 2017), 6.
46 World Bank, 'Summary Update: Security Public Expenditure Review, Somalia Comprehensive
Approach to Security Conference, Mogadishu, December 2017', unpublished document, 1.

THE JOURNAL OF STRATEGIC STUDIES @ 9

After AMISOM expanded beyond Mogadishu in late 2011 it needed
a credible Somali partner force that was disciplined and could assuage
international concerns about accountability and discipline. 47 The United
States therefore supported a private firm that had been working in
Somalia with AMISOM since 2008, Bancroft Global Development, to pilot
a special SNA advanced infantry platoon as proof of concept for what would
become the Danab battalion. Once this was viewed as an operational
success, during 2013 the Somali government and the United States sup¬
ported its expansion to a company and then battalion with Bancroft provid¬
ing training and mentoring, logistics coming via other US contractors, and
salary payments made through PricewaterhouseCoopers, which was con¬
tracted to support the TFG to handle important financial matters. By
July 2014, the first Danab company was fully operational and moved to its
base in Baledogle, north of Mogadishu. The following month it deployed
during AMISOM's Operation Indian Ocean (August-October 2014), which
was intended to clear and then hold about a dozen selected al-Shabaab
strongholds across south-central Somalia. During 2015, the Danab battalion
took shape, including the command and support elements. By recruiting
personnel on the basis of merit from across all south-central Somalia's
regions and clans, paying them consistently, and providing operational
mentoring, the Danab units represented an important symbol of what
military professionalism in Somalia could look like. Importantly, Danab
forces were kept largely separate from the rest of the SNA, which remained
well short of such standards.

Anatomy of failure

External efforts to build an effective SNA were frustrated by three interre¬
lated types of challenges: political, contextual and operational. The next
sections analyse the main problems in each area and how they prevented
the development of an effective SNA.

Political challenges

The key political obstacles were interest asymmetry between international
actors and local Somali elites; insufficient focus on institution-building; and
a lack of coordination among Somalia's international partners.

47 This analysis is based on Danab Battalion Update (Bancroft Global Development briefing,
November 2014) and author's confidential interviews.

10 @ P. D. WILLIAMS

Interest asymmetry

Interest asymmetry among the providers of SFA and the host state autho¬
rities has been identified as one of the general problems with this enterprise
and often results in a relatively small pay-off. 48 In Somalia, the asymmetry
was over the need to achieve national unity, build professional security
forces, and prioritise the fight against al-Shabaab. It produced
a misalignment of political interests between external SFA providers—who
saw the defeat of al-Shabaab and the resurrection of a unified Somali state
as their principal objectives—and the various Somali federal and regional
authorities—who did not always share these priorities. As William Reno
recently put it, ironically, 'state officials who play critical roles in security
assistance programs ... are [also] deeply implicated in the activities and
behaviour that security assistance is meant to change.' 49 The root of this
asymmetry stemmed from Somalia's fragmented national politics, charac¬
terised by mistrust and disunity between the federal and emerging regional
authorities across the country. This reflected the lack of reconciliation
among Somalia's numerous conflicting parties since the civil war
(1988-91) and the country's subsequent dysfunctional political system
wherein elites often had significant incentives to retain the status quo
because they benefited politically and/or economically from the country's
state of insecurity.

Although both the Somali Transitional Federal Governments (2004-09
and 2009-12) and two Federal Governments (2012-17 and 2017-present)
wanted to resurrect a federal state in Somalia, they found themselves in
a complicated relationship with al-Shabaab, which retained political connec¬
tions in key posts in the security sector and was therefore able to infiltrate
Somali security forces. Furthermore, because the Interim Regional
Administrations (IRAs) that emerged across south-central Somalia from
mid-2013 did not always share the goal of creating a strong federal govern¬
ment in Mogadishu they did not always act to defeat al-Shabaab as their
principal threat. Instead, they often become preoccupied with more loca¬
lised and clan-based power struggles (see below). 50 The most recent man¬
ifestation of this dynamic has been the so-called Gulf Crisis, where several
IRAs and members of parliament have willingly accepted bribes from the
UAE to undermine Somalia's President. 51

48 See, for example, Stephen Biddle, 'Building Security Forces and Stabilizing Nations: The Problem of
Agency', Daedalus 146/4 (2017), 126-138; and Stephen Biddle, Julia McDonald, Ryan Baker, 'Small
Footprint, Small Payoff: The Military Effectiveness of Security Force Assistance/ Journal of Strategic
Studies 41/1-2 (2018), 89-142.

49 William Reno, 'The Politics of Security Assistance in the Florn of Africa,' Defence Studies 18/4 (2018),
499.

50 This will be an ongoing issue as international partners decide how much SFA to give directly to
regional forces instead of the SNA.

^International Crisis Group, Somalia and the Gulf Crisis (Report No.260/Africa, 5 June 2018).

THE JOURNAL OF STRATEGIC STUDIES (■+.) 11

In sum, the interests of Somalia's key political leaders were not conducive
to building a set of professional national security services focused on
defeating al-Shabaab. Crucially, these alternative priorities also left Somali
elites unwilling to crackdown on corruption, which massively undermined
the project to build an effective SNA.

In 2017, for example, Transparency International's annual Corruption
Perceptions Index ranked Somalia as the world's most corrupt country for
the eleventh year in a row. 52 Corruption amongst senior commanders and
Somali political elites was another major challenge to building an effective
SNA, not least because the lack of de facto loyalty to the federal government
of most rank-and-file troops meant that their allegiance had to be bought.
As one US official put it back in June 2009, the TFG had 'constantly to
purchase loyalty to motivate militias to fight in its defense.' 53 But chronic
corruption among Somali politicians meant that much of the money osten¬
sibly earmarked for the SNA was stolen.

This corruption came in two main varieties: by elites and rank-and-file
soldiers. Elite corruption involved relatively wealthy and politically powerful
individuals stealing large sums of money and working to oppose strong,
government-led rule of law and a professional security sector in order to
further their narrow economic interests. In July 2012, for instance, the UN
Monitoring Group identified 'pervasive corruption within the transitional
federal institutions.' The group concluded that 'the systematic misappropria¬
tion, embezzlement and outright theft of public resources have essentially
become a system of governance, embodied in the popular Somali phrase
"Maxaa igu jiraa?" ("What's in it for me?").' 54 It estimated that $7 out of every
$10 received by the TFG in 2009 and 2010 never made it into government
coffers and remained unaccounted for.

At the other end of the political pyramid, SNA soldiers were asked to risk
their lives often without receiving meagre salaries, equipment and rations,
or medical care (see below). 55 This encouraged troops to find alternative
sources of income and support, including extorting money or food from
alternative sources, taking multiple jobs in the private sector, obtaining two
or more identification cards to draw multiple salaries, or selling their

“Details of the Index are at www.transparency.org .

53 Wikileak Cable 09ADDISABABA1409, 26 June 2009, §4 and §6.

5A Report of the Monitoring Group on Somalia and Eritrea Pursuant to Security Council Resolution 2002
(2011) (S/2012/544, 13 July 2012) 7.

55 Since the new FGS was established in September 2012, SNA troops were supposed to receive
a monthly wage of $260 ($100 from the government, $60 for rations, and $100 from a donor).
Due to financing constraints, by 2016, the average SNA soldier was scheduled to receive: $50 salary
from the Federal Government, $60 for food rations ($30 to be received in cash, $30 to be received in-
kind), and $100 donor financed stipend per month for the average soldier. Between 2013 and 2016,
over 80% of SNA expenditures went on wages and rations (with the remainder going on operations).
Author's confidential interview, March 2017. See also UNSOM and World Bank, Somalia Security and
Justice Public Expenditure Review (UNSOM and World Bank, January 2017) xii, 24 and 59.

12 @ P. D. WILLIAMS

equipment. During the TFG years, it was not uncommon for soldiers to sell
some of their limited ammunition to buy food and khat, and then abscond
when they had used the rest. 56 This was corruption but of a qualitatively
different kind and left SNA troops with little incentive to fight. As the Mayor
of Mogadishu had quipped in 2009, he had 6000-7000 TFG soldiers on
payday but only 2000-3000 when it was time to fight. 57 These forms of
corruption also turned elements of the local population against the SNA and
left the troops vulnerable to exploitation.

AMISOM was so concerned about the potential for the TFG troops to sell
their equipment and the weak or non-existent accountability mechanisms
that the mission concluded the TFG should not be permitted to stockpile
internationally provided military supplies, such as fuel and ammunition. 58
AMISOM also decided to oversee 'the storage and dispensation' of the TFG's
arms and ammunition and rations 'in order to stem the Government corrup¬
tion that had surfaced in previous years.' 59 This led TFG commanders to
complain that they were being undermined by AMISOM's unwillingness to
provide them with sufficient ammunition. 60 Nevertheless, AMISOM was
unable to prevent ammunition leaking 'from the custody of Government
and militia commanders to the illicit market.' 61 By 2011, the UN Monitoring
Group thought TFG and aligned forces sold between one-third and one-half
of their ammunition. 62

Finally, opportunities for both elite and rank-and-file corruption were
increased because salary payments to the SNA were usually made in cash.
This eventually prompted donors to demand that the Somali government
use a PricewaterhouseCoopers mechanism to disburse funds using the
mobile phone network; once the soldier's identity and bank details were
verified. The problem here was political not technical since Somalis regularly
use such systems for all sorts of financial interactions. Specifically, electronic
payment systems were hampered by SNA commanders who would lose
their cut if they could no longer directly disburse cash payments to the
troops. Today, the systems and controls in place to oversee the payment
process remain extremely weak, which encourages leakage. 63 In late 2017,
a pilot program was tested on the SNA battalion that guards Villa Somalia
but it has yet to spread to the rest of the army. 64

56 Security Sector Assessment (June 2011) 26.

57 Wikileak Cable 09NAIROBI1798, 25 August 2009, §5.

5S Somalia Security Sector Assessment (January 2010) §16a9b10.

59 Report of the Monitoring Group on Somalia and Eritrea pursuant to Security Council Resolution 1916
(2010) (S/2011/433, 18 July 2011), §133. See also Wikileak Cable 09NAIROBI1520, 15 July 2009, §7.
60 Wikileak Cable 09NAIROBI1370, 2 July 2009, §6.

61 SEMG Report, 18 July 2011, §135.

62 lbid, Annex 5.1, 231.

63 UNS0M and World Bank, Somalia Security, 104 and x.

^Communication, US official, 10 January 2018.

THE JOURNAL OF STRATEGIC STUDIES (■+.) 13

Insufficient focus on institutions

A second political problem was trying to build a national army without
dedicating enough effort and resources to develop the relevant institutions
and policy frameworks. As a result, the focus often became killing particular
targets rather than building sustainable and effective institutions. 65 But
institutions are crucial for an effective army, not least to develop its strategic
vision and force structure, assign its tasks, ensure sustainable procurement
and logistics support, as well as educate, manage and care for its personnel
(and their dependents). Until April 2017, however, Somalia lacked the basic
building blocks of a national security architecture let alone policy or opera¬
tional frameworks into which to plug the various international SFA pro¬
grammes. Instead of joined-up efforts to develop a security architecture,
national security strategy, military doctrine, and educational and procure¬
ment institutions to support newly recruited soldiers, the various 'train and
equip' efforts were left to fester in an institutional vacuum. To give one
example, the very limited human resources capacity in the Somali Ministry
of Defence—which by mid-2017 had only about 60 civilian personnel,
including support staff—made it difficult to absorb and coordinate the
assistance received let alone manage a national army. Consequently, even
when specific SNA troops and units were trained and equipped there was
no institutional infrastructure to manage them, assign them appropriate
tasks, or get them operating in the field. In sum, 'train and equip' initiatives
alone could not generate an effective army in the absence of such
institutions.

Uncoordinated partners

The third key political problem was the lack of coordination between the
SNA's external partners. Ideally, multiple SFA providers would follow similar
(or at least compatible) political objectives and military doctrine, deliver
interoperable equipment and training on maintaining donated equipment,
use standardised training programmes, and base them on an assessment of
the SNA's needs. Indeed, faced with a plethora of external initiatives, in
2012, the new Federal Government asked partners to organise a single 'door
to knock on' to coordinate activities in the security sector. 66 In reality,
however, the SNA received uncoordinated external support for more than
a decade. 67

65 See Reno, 'The Politics', 500.

“'Letter dated 19 April 2013 from the Secretary-General addressed to the President of the Security
Council' (S/2013/239, 19 April 2013) Annex §43.

67 See, for example, the similar conclusions reached in Somalia Security Sector Assessment
(January 2010); AU, Report on TFG (November 2011); Security Sector Assessment (June 2011); and
Brief Report (October 2011).

14 @ P. D. WILLIAMS

Most SFA was provided in ways that aligned with the partner's own vision
and strategic interests. Some Somalis even saw this as a deliberate attempt
to fragment their fledgling security sector and noted how clandestine
military operations run by Ethiopia, Kenya and the United States—some¬
times involving Somali actors—all complicated the issue. 68 On the other
hand, without Somalia's political elites agreeing on a shared vision for their
national security architecture and forces, SFA providers had no stable frame¬
work into which to plug their assistance. 69

Even when partner agendas converged around similar broad objectives
they were rarely well coordinated. Or rather, select groups of donors man¬
aged some coordination, notably the US, UK and EU, but other states
including the UAE, Ethiopia, Kenya and to a lesser extent, Turkey, would
conduct their own disconnected projects. In mid-2016, the 'S6' group was
formed to bring the external SFA initiatives of Turkey, the UAE, EU, UN, US
and UK into line. But it had limited success and was quietly abandoned the
following year and replaced by the Comprehensive Approach to Security,
which encouraged partners to adopt an agreed division of labour across the
five strands of (1) enabling effective AMISOM operations; (2) developing
Somalia's national security institutions; (3) stabilisation efforts; (4) countering
and preventing violent extremism; and (5) establishing an effective coordi¬
nation mechanism. 70

Take training programmes as one example of poor coordination. Since
2009, they were provided in several different languages, including English,
French, Swahili, Amharic and Arabic, leaving Somali trainees with diverse
military skills and doctrines only to find another form of military training
skills and doctrine back home. 71 Unfortunately, partners did not keep track
of many of their trainees, which led to wide-ranging numbers up to some
100,000 soldiers having received training, some of them probably many
times over. 72 During the late 2000s, one expert concluded that 'most of
the units that were trained defected.' 73 Even by 2013, the principal training
facility in Mogadishu—Jazeera camp—lacked adequate billets, sanitation,
and manoeuver space for exercises, apparently causing some recruits to
leave. 74 More fundamentally, training alone was not sufficient to build an

68 lnterview, FGS official, Djibouti, 27 February 2016.

69 Security Sector Assessment (June 2011), 33.

70 As set out in the London Somalia Conference Security Pact, 11 May 2017, https://www.gov.uk/
government/publications/london-somalia-conference-2017-security-pact .

71 Brief Report (October 2011).

72 Telephone interview, UN Official, 7 September 2017.

73 Stig Jarle Hansen, At-Shabaab in Somalia (London: Hurst 2013), 96.

74 Roger Carstens, 'Analog War: How to rid Somalia of al-Shabab once and for all - in six (not-so) easy
steps', Foreign Policy.com, 7 October 2013. The Jazeera Training Centre, located southwest of the
Mogadishu International Airport, had trained Somali troops since at least 2010. By 2012, there were
45 instructors from Burundi, Kenya and Uganda, 25 Somali instructors, and supplementary staff from
the AMISOM force headquarters.

THE JOURNAL OF STRATEGIC STUDIES (■+.) 15

effective army. As the 2009 Somali security sector assessment suggested, it
'may even be counterproductive if it is not followed up by a sustainable
equipment/supply program and payment of salaries. If these requirements
are not met, the risk of desertion/defection of trainees upon return to
Somalia is high.' 75 In this sense, Matt Bryden was correct to argue that
paying salaries was ultimately 'much more important in building
a professional force than obtaining new weapons and equipment.' 76 The
2015 Guulwade Plan sought to address some of these problems by coordi¬
nating the assistance packages that its forces received around the new FGS
training plan, which involved establishing sector and combined training
centres, mentors, and a schedule for training priorities for the SNA. 77 From
2015, the US-led Military Coordination Cell in Somalia did enable better
coordination at the technical and operational level but it was not until early
2017 that the Joint Training Symposium of partners made an important
political breakthrough and adopted a Somali-owned training programme
based on NATO standards. 78

Contextual challenges

Somalia's recent history and cultural dynamics also stymied international
attempts to build an effective SNA in two major respects: first, the unhelpful
legacies of two decades of state collapse, and second the multiple chal¬
lenges raised by the country's clan politics.

Legacies of state collapse

Between 1991 and 2000 Somalia had no central government, and from 2000
to 2012 its transitional governments commanded little loyalty and con¬
trolled little territory beyond parts of Mogadishu. This extended period of
state collapse led most locals to find alternative security providers. Ken
Menkhaus famously described a situation of 'governance without govern¬
ment,' whereby various non-state actors and informal systems of adaptation
delivered services 'in the prolonged absence of a central government.' 79
Once international efforts started to resurrect a Somali federal government
through the TNG and TFG in the 2000s, this produced a hybrid political
order in which official governing institutions had to compete or negotiate
for authority with other non-state entities to play the role of key security

75 Somalia Security Sector Assessment (January 2010), §16a9b8j.

76 IRIN, 'Somali Security'.

77 Guulwade Plan, 5.

78 SNA presentation to AMISOM Joint Review team, Mogadishu, 23 May 2017.

79 Ken Menkhaus, 'Governance without Government in Somalia', International Security 31/3 (2006/07),
74.

16 @ P. D. WILLIAMS

provider. More often than not, it was clan militias that protected local
populations across most of the country. 80

This political context generated several symptoms that made it extremely
difficult to build an effective SNA. First, it was tough to build a national army
while simultaneously trying to re-establish a federal government. Until
September 2012, Somalia had only transitional national authorities, which
as noted above controlled very little territory and provided almost zero
public services. Between 2013 and 2017, the often conflict-inducing process
of establishing the new IRAs across south-central Somalia increased ten¬
sions, especially when external partners attempted to support regional
forces beyond the Federal Government's control. 81 This stemmed from the
lack of reconciliation that had taken place between these actors since the
end of the civil war in 1991 82

The second, related problem was the widespread and deep distrust
among many Somalis beyond the Banadir region that the Federal
Government would be an impartial actor or could deliver genuine security.
This bred a strong inclination towards continued reliance on non-state
actors by both regional administrations and many ordinary Somalis, which
in turn made it difficult to attract sufficient numbers of recruits to the SNA.
Indeed, for those actors that saw a strong federal government as a potential
threat, the idea of giving it an effective army was a recipe for disaster and
marginalising other groups.

Third, this situation generated strong loyalties to actors beyond the
Federal Government, which complicated any attempt to forge a unified
and coherent command and control system throughout the SNA.

Fourth, prolonged state collapse meant that when the Federal
Government did establish new institutions they were dysfunctional and
hence sometimes simply reaffirmed people's suspicions of government
incompetence and the need for alternatives. To take one example, without
sustainable revenue it would be impossible to build an effective SNA. Yet
Somalia had completely dysfunctional financial institutions such as revenue
authorities, customs and the Central Bank. Consequently, the overwhelming
majority of Somalis did not pay taxes, which produced very limited govern¬
ment revenues, much of which was embezzled because of elite corruption.
Since the Central Bank did not perform its fiduciary responsibilities interna¬
tional agencies had to rely on Xawala or the PricewaterhouseCoopers
mechanism and could not stop many illicit financial transactions involving

80 Report of the Monitoring Group on Somalia pursuant to Security Council resolution 1724 (2006) (S/2007/
436, 18 July 2007) 11.

81 Between 2013 and 2016, four IRAs were established across south-central Somalia: Jubbaland,
Southwest Somalia, Hirshabelle and Galmudug, each of which sought to develop their own regional
forces not all of which were integrated into the SNA.

82 See, for example, Tobias Hagmann, Stabilization, Extraversion and Political Settlements in Somalia
(Nairobi: Rift Valley Institute, 2016).

THE JOURNAL OF STRATEGIC STUDIES (■+.) 17

unregistered private and even foreign governments. 83 In such circumstances
there was little prospect of generating sustainable revenues let alone estab¬
lishing a workable procurement system for a military.

A fifth legacy problem was the long period without a functioning military
system. Combined with Somalia's youthful population, this left many young
potential SNA recruits confused about the meaning of a federal state and
how it was supposed to function. 84 It also left behind a missing generation
of officers available to the armed forces. As one senior TFG commander put
it in 2011, 'the most critical gap in the existing forces was the lack of young
officers and junior officers. Most officers are old: "less than 50" are younger.
The older officers find it difficult to lead the young soldiers, physically and
mentally.' 85 Poor education and literacy levels were also a problem that
made some military speciality categories difficult to populate. 86

Finally, it should be noted that before the IRAs were established between
2013 and 2016, there was no political entity at the sub-federal level that had
official legitimacy to discuss the politics of national security. It was only after
the creation of these entities that legitimate answers to the questions of
security governance could be provided. Before that, it had been
a Mogadishu-focused process, which fuelled critics in the regions.

Clan politics

Although Somalia's political elites do not automatically make decisions
based on their clan identity, clan politics cast a large shadow over the
country's security sector and significantly complicated the goal of building
an effective national army. In his careful study of the Somali Armed Forces,
Colin Robinson concluded 'Clan loyalties exceeding loyalties to the center is
the principal obstacle to rebuilding an effective national army for Somalia.' 87
'Clan ties', he argued, 'trump virtually everything, business interests some¬
times excluded.' 88

The roots of major clan divisions and agendas rupturing the SNA can be
traced back to Siad Barre's efforts to consolidate his rule during the 1980s. 89
This came after an aborted coup in 1978, when Barre responded by loading

83 Security Sector Assessment (June 2011), 30-31.

M Brief Report (October 2011).

85 Cited in Robinson, 'Revisiting', 246.

86 Paul R. Camacho and Ibrahim Mohamed Abukar, Somalia's Security: The Reconstruction of the Somali
National Army (Mogadishu: Center for Policy Analysis and Research February 2017), 8. See also
Mubarak, 'Somali Military'.

87 Robinson, 'Revisiting', 245. A similar point is made by Camacho and Abukar, Somalia's Security, 7.

88 Robinson, 'Revisiting', 246. On the importance of business interests, see Aisha Ahmad, 'The Security
Bazaar: Business interests and Islamist power in civil war Somalia', International Security 39/3 (2014/
15), 89-117.

89 Before 1976, the armed forces sought a balance between the major clans, even specifying the
proportional number of recruits from each district. See Robinson, 'Revisiting'.

18 @ P. D. WILLIAMS

the armed forces with his Marehan and related Darod officers. Clan
dynamics undermined the SNA ever since. With the implosion of the central
government in 1991 clan dynamics became even more important as most
individuals turned to their clans, sub-clans or sub-sub clans to provide for
their security.

By the time of the first TFG administration, the struggle over Mogadishu
was interpreted by many Somalis as primarily a conflict involving a Darod-
led TFG trying to re-assert Darod dominance over a primarily Flawiye city.
Following the selection of the second TFG administration in January 2009,
the Flawiye/Abgaal and Hawiye/Flabir Gedir/Ayr increased their power over
the military sphere around Mogadishu. 90

Clan politics also caused problems as AMISOM and the TFG forces moved
beyond Mogadishu from late 2011. With al-Shabaab retaining control of
much of south-central Somalia, this left relatively few areas from which the
SNA could recruit new soldiers. Specifically, the Banadir region around
Mogadishu was dominated by particular clans and sub-clans, which were
then directly reflected in the composition of new recruits entering the army.
One internal AMISOM analysis concluded that the 'monster of clan' infected
the whole enterprise. 'Clan identity within the forces,' the AU noted, 'is the
stumbling block to all efforts in organizing a national military force.' 91 The
practical consequence of the preponderance of certain clans was that the
ostensibly 'national' army was perceived by most citizens outside
Mogadishu as a partisan force dominated by particular sub-clans. When
those forces operated outside their areas of influence, such as Lower
Shabelle, locals viewed them with suspicion.

Clan politics also had debilitating effects on achieving a unified com¬
mand and control structure within the SNA both because rank-and-file
troops often displayed loyalty to clan (and other actors, including previous
warlords) rather than the federal government, and because sometimes clan
politics complicated relationships between some senior Ministry of Defence
personnel and the SNA. With most SNA battalions and brigades largely
organised around clan lines, this reduced formal interaction and collabora¬
tion between them. In some cases, clan conflicts stopped collaboration
between battalions within the same brigade. 92 As a 2015 UNSOM report
observed, the subsequent dynamic within the units saw most of them
revolve around one strong leader with senior personnel from the same
clan; if company and platoon commanders were not of the same clan,
however, command authority quickly broke down. 93 Most of these com¬
manders had little formal training but considerable militia experience. This is

“'Leaked SEMG letter, S/AC.29/2014/COMM.13 (0C.8), 6 February 2014/ 5 cited in UNSOM, The SNA, 15.
91 AU, Report on TFG (November 2011), 9 and 1.

92 lbid, 6.

“UNSOM, The SNA, 6.

THE JOURNAL OF STRATEGIC STUDIES 0*0 19

why some experts concluded that the importance of sub-clan dynamics
dictated that painstaking local-level negotiations were required to find
command solutions. 94

Clan dynamics also complicated the payment of salaries for SNA
troops. Not only have clan leaders that provided fighters to serve in
the SNA demanded kickbacks on the soldiers' pay but clan favoritism
within the SNA has meant troops outside of Mogadishu and Middle and
Lower Shabelle regions have received little if any salary support. 95

Finally, clan politics undermined the potential for the new SNA to act
as a unifying symbol for Somalia. Multiclan SNA units were potentially
important symbols of national unity. 96 But they also represented a threat
to the existing clan influence within Somalia's security sector. As
a consequence, they existed in only a couple of SNA brigades and the
Danab advanced infantry battalion, and only in the latter case were
recruits consistently selected on merit across clan. An SNA largely segre¬
gated along clan lines could not act as much of a unifying force.

Operational challenges

The main operational challenges involved building an army while simul¬
taneously fighting a war against a deadly and adaptable opponent, the
complexities of military integration in a country with so many armed
groups, and the huge capability gaps afflicting the SNA.

Reforming while fighting

Most SSR programmes take place after a war or during the process of forging
a political settlement. In Somalia, however, building the SNA occurred during
wartime, in the absence of a peace process, and until very recently, without
an agreed national security architecture, strategy and force structure. As the
UN's senior military advisor in Somalia acknowledged in May 2013, this
produced 'a real tension trying to train these [SNA] forces when they are at
war' because 'no one has the luxury of pulling them out; they are essential in
the fight against Al-Shabab.' 97 This raised questions about how applicable
standard SSR guidelines and principles were for Somalia. US guidelines for

94 Matt Bryden and Jeremy Brickhill, 'Disarming Somalia: Lessons in Stabilisation from a Collapsed State',
Conflict, Security & Development 10/2 (2010), 239-62.

95 Report of the Monitoring Group on Somalia and Eritrea pursuant to Security Council resolution 2244
(2015): Somalia (S/2016/919, 31 October 2016) Annex 2.2, §8.

96 0n the important symbolic roles of national armies in the unifying war-torn territories see Roy
Licklider, 'Introduction', in Roy Licklider (ed.), New Armies from Old (Washington DC: Georgetown
University Press 2014).

97 'Somali Security Sector Reform', IRIN, 13 May 2013, http://www.irinnews.org/analysis/2013/05/13/
somali-security-sector-reform .

20 @ P. D. WILLIAMS

successful SSR, for instance, emphasise that implementing it during political
instability and without national consensus on the vision and structures guid¬
ing the sector is a recipe for failure. 98 Similarly, Somalia lacked most of the
'fundamental core elements' identified by the AU for promoting SSR. 99 Some
analysts therefore warned that SFA in Somalia would be vulnerable to high
waste and potentially counterproductive. 100

But there were few options: the SNA had to be built while it was fighting. The
initial plan was to take the fight to al-Shabaab, win the war and then interna¬
tional partners could focus on professionalising the Somali security forces.
A rough but uncoordinated division of labour emerged. The US focused on
building the Danab battalion while Ethiopia concentrated on the Gedo, Bakool
and Hiraan regions, Kenya operated in the Jubba regions, whereas Turkey, the
UAE and EU remained in Mogadishu.

Other operational challenges emerged as well. During 2009 when AMISOM
and the TFG fought against al-Shabaab over parts of Mogadishu, the lack of
even basic equipment such as uniforms left AMISOM commanders concerned
about potential problems in distinguishing non-uniformed TFG troops and
allied militias during combat. Other deficiencies led TFG commanders to
worry that even when they captured new territory their 'front line' troops
were not always committed to hold it. 101 In mid-2011, at the height of the
battle for Mogadishu, the lack of basic communications equipment and skills
on the Somali side meant that there was only 'limited coordination' between
AMISOM and Somali forces, which made 'it difficult to conduct joint planning
and operations to contain insurgent groups fighting against the TFG.' 102

To make matters more difficult, Somalia was what one contractor working
with the SNA called a 'distributed battlefield,' that is, a large territory with few
troops. 103 In such terrain, emphasis should be placed on small battalions
(rather than brigades) as the focal point of the army and success rested on
the ability to operate with 'small unit tactics, mobile strike teams, civil affairs,
and simplified logistics.' But these were tools that would take time, resources
and training to master, none of which were available to the SNA at the time.

Problems of military integration

Another practical problem was whether and how to build the new SNA
through an official process of military integration—the process of bringing
individuals 'into the new military in positions similar to the ones they

98 US, Security Sector Reform (US AID, Department of Defense, Department of State, February 2009),
https://www.state.gOv/documents/organization/115810.pdf .

"See Policy Framework on Security Sector Reform (Addis Ababa: AU 2013) Section D.

100 UNSOM, The SNA, 21; author's communication, Colin Robinson, 6 August 2017.

101 Wikileak Cable 09NAIROBI1962, 18 September 2009, §2.

102 Security Sector Assessment (June 2011), 6.

103 Carstens, 'Analog War'.

THE JOURNAL OF STRATEGIC STUDIES @ 21

occupied in prior organizations'. 104 In Somalia, SFA providers confronted
a hybrid political order that resembled what Alice Hills called a 'security
arena,' in which a multitude of armed groups competed for dominance,
rather than an institutionalised 'security sector'. 105 In this environment,
some degree of military integration was necessary for building a national
army. And as discussed above, since 2008 the SNA was cobbled together
out of various clan forces, former Islamic Courts militias and subsequent
rebels, old SNA 'grey soldiers,' and other ostensibly government-aligned
groups of fighters.

Sometimes this was done formally as a way to build greater political unity
among armed factions, as in the Djibouti Agreement (2008). But on most other
occasions it happened by default or to appease particular leaders who retained
personal command of 'their' fighters and were bought into the SNA fold.

In this sense, military integration wasn't just a peacebuilding strategy; it was
also a potential means to wage war more effectively against al-Shabaab and
provide an exit strategy for AMISOM. But it came with some considerable risks,
not least of rogue commanders, al-Shabaab infiltration, and the absence of an
overarching peace process to define which actors to integrate and on what
terms. Hence the SNA remained a messy amalgamation of multiple armed
groups with a distinct lack of unified command and control. The problems of
integration intensified further from 2013 with the process of establishing
Somalia's Interim Regional Administrations and when their subsequent, largely
clan-based regional forces assumed a more important status. This presented
a dilemma of whether to support their continued existence outside of the SNA
structure in the hope they would help degrade al-Shabaab or try and officially
integrate them into the SNA chain of command and dilute their clan affiliations.
An operational readiness assessment of Somalia's regional forces was supposed
to be completed by the end of 2018 but remains incomplete.

Military integration was thus a deeply political process in Somalia, which
no amount of technical SFA could resolve. In particular, either for clan-
related or political reasons many of the armed groups beyond Mogadishu
distrusted attempts to centralise control of the Somali armed forces. As one
scholar correctly summarised, military integration 'is, first and foremost,
a local political problem, not a matter of technical expertise that local actors
lack and whose secrets international actors must therefore impart'. 106 And as
analysed above, the interest asymmetry, the lack of reconciliation between
major armed groups, and the absence of a national security architecture

104 Licklider, 'Introduction', 3.

105 Alice Hills, 'Security Sector or Security Arena? The Evidence from Somalia', International Peacekeeping
21/2 (2014), 165-80. In 2016, ACLED identified 150 distinct armed groups in Somalia. See https://
www.acleddata.com/dashboard/#706.

106 Ronald Krebs, 'So What?', in Licklider (ed.), New Armies from Old, 255.

22 @ P. D. WILLIAMS

until April 2017 meant that the politics of Somalia's military integration
process remained unconducive to building an effective national army.

Capability gaps

The final key operational challenge was the SNA's daunting list of capability gaps.
As one recent independent study concluded, 'The list of needs is so fundamental
that it is no exaggeration to suggest that Somalia is building its army from the
very foundation.' 107 Even when launching its Guulwade Plan in 2015, the Federal
Government acknowledged a long list of technical and infrastructural shortages
afflicting the SNA. 108 The principal operational gaps were mobility, especially
armoured vehicles but also 'technicals,' pickups, troop-carrying trucks, and spe¬
cialist vehicles such as ambulances, water trucks, fuel bowsers, and recovery
vehicles; all types of ammunition 109 ; heavy weapons and the ability to maintain
and repair them 110 ; communications equipment, including military radios 111 ; and
field defences, especially force protection for defended positions and forward
operating bases. There were also very limited counter-IED capabilities, which was
reflected in the fact that between 2015 and 2017 the SNA suffered nearly twice as
many fatalities from IED attacks as AMISOM. 112 The lack of small arms and light
weapons is particularly interesting given the corruption discussed above.
According to the UN Monitoring Group's 2015 report, following the partial lifting
of the UN arms embargo in 2013 the Somali government received over 17,500
weapons. 113 And yet the 2017 Operational Readiness Assessment of the SNA
revealed that only 70% of troops possessed any weapons and only an average of
195 service rifles in each battalion.

The SNA also still had major shortages and problems related to training,
logistics capacity, vehicle maintenance facilities, arms and ammunition storage
facilities, medical support (especially facilities and resources to treat and stabilise
patients forward in the sectors), 114 and adequate barracks to accommodate and
maintain good control over its troops. The combined effect of these gaps was not
only to reduce military effectiveness but also to undermine morale.

107 Camacho and Abukar, Somalia's Security, 11.

108 Guulwade Plan, 2-8. Similarly long lists of equipment and infrastructural shortages can be found in
Security Sector Assessment (June 2011) and Brief Report (October 2011).

109 Although note that the UN Monitoring Group on Somalia concluded the SNA should have received
almost 9 million rounds of ammunition. Report of the Monitoring Group on Somalia and Eritrea
pursuant to Security Council resolution 2182 (2014): Somalia (S/2015/801, 19 October 2015), §136.

110 ln 2011, AMISOM estimated that 80% of SNA weapons belonged to warlords, clans and individuals
rather than the Federal Government. AU, Report on TFG (November 2011), 7-8.

ln For years, the TFG troops had to rely on commercial mobile phone networks for communications or
walkie-talkie handsets. Somalia Security Sector Assessment (January 2010), §4ag and Brief Report
(October 2011).

"interview, UNMAS official, Mogadishu, 24 June 2018.

' n SEMG Report, 19 October 2015, §136.

114 By late 2011, the Martini Hospital in Mogadishu was the SNA's only medical facility. It had a 20-bed
capacity, one volunteer doctor and about 12 nurses. AU, Report on TFG (November 2011), 5.

THE JOURNAL OF STRATEGIC STUDIES @ 23

It will suffice to end by briefly noting that the SNA lacked arguably the most
basic capability an army should possess: the ability to identify its personnel and
stop desertion. Even today, however, the SNA is not able to stop its soldiers
deserting or absconding. 115 The reasons varied: some left because their real
loyalty lay elsewhere, for others it was lack of regular pay, food or other poor
conditions of service. But the problem was compounded by the lack of
a comprehensive biometric identification system, which is still not complete.
Instead, the FGS, the US and UK, and the UAE all held SNA identification registries
but none were fully comprehensive nor were they aligned. 116 Unsurprisingly,
therefore, the SNA suffered from ghost soldiers on its roll and stipends going to
people who were dead, permanently injured, old persons, women partners,
orphans, and disabled. 117 Some SNA personnel and their political masters used
this state of confusion to engage in corruption and fraud; al-Shabaab exploited it
in order to infiltrate the SNA to prosecute attacks on government installations and
personnel.

Conclusion

Without an effective Somali army, little military pressure could be put on al-
Shabaab and AMISOM's exit strategy was effectively stalled. Yet considerable
time, money and effort by over a dozen international partners failed to build such
a force. The explanation for this failure lies in three interrelated sets of political,
contextual and operational challenges. Numerous unhelpful legacies of two
decades of state collapse and the complexities of Somalia's clan dynamics
certainly represented an almost uniquely difficult context in which to try and
build an effective national army. But with sufficient political will and unity of
purpose among Somali leaders and international partners these constraints need
not have guaranteed failure. However, key Somali elites lacked the requisite
political will to build professional, national security forces. This interest asymme¬
try between the main priorities of international partners and the Somali autho¬
rities badly undermined attempts to build an effective SNA. The inability to deal
with pervasive corruption and forge a consensus among Somalia's political elites
around a shared vision of the national security architecture, how to finance it, its
decision-making mechanisms, and how best to prioritise the fight against al-
Shabaab, left both AMISOM and international partners with an impossible task.
Specifically, it stymied the establishment of the kinds of security institutions that
a national army needs to function and left international partners without
a settled framework into which to plug their SFA. The result was various partner
projects trying to build an army in separate pieces. Nevertheless, Somalia's

115 For a list of SNA withdrawals and salary-related incidents between September 2015 and
August 2016, see SEMG Report, 31 October 2016, Annex 2.6.

116 FGS/DFID, Securing Payroll, 41. See also 5£MG Report, 31 October 2016, Annex 2.3.

117 AU, Report on TFG (November 2011), 8.

24 @ P. D. WILLIAMS

partners could have done a much better job of coordinating their disparate SFA
programmes. To make matters worse, even if the political context had been more
conducive to building such institutions, the SNA had to be built almost from
scratch while simultaneously fighting a war against a deadly, adaptable and
knowledgeable foe. And here the SNA suffered from almost every conceivable
technical and institutional capability gap from weapons to vehicles, barracks to
medical support and identification systems to salaries.

The current national security architecture endorsed in the London
Security Pact (May 2017) and the Somali Transition Plan (drafted in
March 2018) provide the best framework thus far on which to forge the
necessary political consensus. But their implementation will require genuine
buy-in from Somalia's regional authorities and political leaders. They must
be willing to confront the spoilers within the country's political establish¬
ment who wish to scupper the development of professional national secur¬
ity forces. This requires Somalia's leaders to take those difficult political
decisions about governance and financing as soon as possible.

Achieving such buy-in will require a fundamental change of attitude and
behaviour by Somalia's leaders, which is unlikely. One attempt to encourage
such a change is to impose more conditions on SFA. In December 2017, the
United States adopted this approach when it cited concerns about corruption
and suspended its stipends and other forms of security assistance to SNA units,
with the exception of Danab and the presidential guard. To restart US assis¬
tance, Somali leaders must stem corruption and deliver on their promises to
implement the details of the new national security architecture. Another way to
encourage greater buy-in is to improve international coordination by adopting
the lead-nation model of SSR. This was used to good effect in Sierra Leone
where the British-led International Military Assistance Training Team forged
a new and effective army. 118 Neither approach is guaranteed to work, however.
If elite buy-in proves elusive, armed forces across Somalia will remain fragmen¬
ted and multiple external actors will likely continue to build separate, localised,
probably clan-based forces (with Danab as the multiclan and meritocratic
exception). At best, such a balkanised approach might reduce the al-Shabaab
threat in particular areas but it will not produce a professional national army or
sustainable national security institutions any time soon.

Acknowledgments

I would like to thank Lesley Warner for her constructive comments on an earlier draft
of this article. This work was supported by the Elliott School of International Affairs at
the George Washington University via a SOAR award.

118

’See Peter Albrecht and Paul Jackson, Securing Sierra Leone, 1997-2013 (London: Routledge 2015).

THE JOURNAL OF STRATEGIC STUDIES @ 25

Disclosure statement

No potential conflict of interest was reported by the author.

Funding

This work was supported by a SOAR Award from the Elliott School of International
Affairs at the George Washington University.

Notes on contributor

Paul D. Williams is Associate Professor in the Elliott School of International Affairs at
the George Washington University. His books include Fighting for Peace in Somalia: A
history and analysis of the African Union Mission (AMISOM), 2007-2017 (Oxford
University Press, 2018) and War and Conflict in Africa (Polity, 2nd edition, 2016).

ORCID

Paul D. Williams ® http://orcid.org/0000-0002-9716-1294

Bibliography

Ahmad, Aisha, 'The Security Bazaar: Business Interests and Islamist Power in Civil War
Somalia', International Security 39/3 (2014/15), 89-117.

Albrecht, Peter and Paul Jackson, Securing Sierra Leone, 1997-2013 (London:
Routledge 2015).

AU, Policy Framework on Security Sector Reform (Addis Ababa: AU 2013), Section D.

Biddle, Stephen, 'Building Security Forces and Stabilizing Nations: The Problem of
Agency', Daedalus 146/4 (2017), 126-38. doi: 10.1162/DAED_a_00464.

Biddle, Stephen, Julia McDonald, and Ryan Baker, 'Small Footprint, Small Payoff: The
Military Effectiveness of Security Force Assistance', Journal of Strategic Studies 41/
1-2 (2018), 89-142. doi: 10.1080/01402390.2017.1307745.

Bryden, Matt and Jeremy Brickhill, 'Disarming Somalia: Lessons in Stabilisation from
a Collapsed State', Conflict, Security & Development 10/2 (2010), 239-62. doi:
10.1080/14678801003665992.

Camacho, Paul R. and Ibrahim Mohamed Abukar, Somalia's Security: The
Reconstruction of the Somali National Army (Mogadishu: Center for Policy
Analysis and Research February 2017).

Carstens, Roger, 'Analog War: How to Rid Somalia of al-Shabab once and for All - In
Six (Not-So) Easy Steps', Foreign Policy.com, 1 Oct. 2013. https://foreignpolicy.com/
2013/10/07/analog-war/.

Hagmann, Tobias, Stabilization, Extraversion and Political Settlements in Somalia
(Nairobi: Rift Valley Institute 2016).

Hansen, Stig Jarle, Al-Shabaab in Somalia (London: Hurst 2013).

Hills, Alice, 'Security Sector or Security Arena? the Evidence from Somalia', International
Peacekeeping 21/2 (2014), 165-80. doi: 10.1080/13533312.2014.910400.

IRIN, 'Somali Security Sector Reform', IRIN, 13 May 2013. http://www.irinnews.org/
analysis/2013/05/13/somali-security-sector-reform.

26 @ P. D. WILLIAMS

Krebs, Ronald, 'So What?', in Roy Licklider (ed.), New Armies from Old (Washington DC:
Georgetown University Press 2014), 259-268.

Licklider, Roy, 'Introduction', in Roy Licklider (ed.), New Armies from Old (Washington
DC:Georgetown University Press 2014), 1-12.

Lunn, Jon, Somalia: February 2017 Update. House of Commons Briefing Paper
No.7298. 16 Feb. 2017.

Menkhaus, Ken, 'Governance without Government in Somalia', International Security
31/3 (2006/07), 74-106.

Mubarak, Mohamed, 'Somali Military Has More Problems than Lack of Guns', African
Arguments, 28 Feb. 2014. http://africanarguments.org/2014/02/26/somali-military-
has-more-problems-than-lack-of-guns-by-mohamed-mubarak/.

Reno, William, 'The Politics of Security Assistance in the Horn of Africa', Defence
Studies 18/4 (2018), 498-513. doi: 10.1080/14702436.2018.1463819.

Robinson, Colin, 'Revisiting the Rise and Fall of the Somali Armed Forces, 1960-2012',
Defense & Security Analysis 32/3 (2016), 237-52. doi: 10.1080/14751798.2016.1199122.

Tokyay, Menekse, 'Turkish Military Base in Somalia', Arab News, 17 Aug. 2017. http://
www.arabnews.com/node/1145846/middle-east.

UNSOM and World Bank, Somalia Security and Justice Public Expenditure Review
(Mogadishu: UNSOM and World Bank January 2017).

US, Commander's Handbook for Security Force Assistance (Kansas: Joint Center for
International Security Force Assistance 14 July 2008).

US, Security Sector Reform (US AID, Department of Defense, Department of State
February 2009). https://www.state.g0v/documents/organization/115810.pdf.

Williams, Paul. D., Fighting for Peace in Somalia: A History and Analysis of the African
Union Mission (AMISOM), 2007-2017 (Oxford: Oxford University Press 2018).

Williams, Paul D and Abdirashid Hashi, Exit Strategy Challenges for the AU Mission in
Somalia (Mogadishu: HIPS February 2016).